Incentives boost process measure performance, but don’t reduce outcome or spending disparities.


Jan. 13—Pay-for-performance (P4P) strategies could help reduce healthcare disparities and improve quality of care for disadvantaged individuals--if the incentives are strong enough, a new Health Affairs report suggests.

Harvard Medical School researchers studied changes in process measures, outcomes, and spending among enrollees whose physicians participated in the Blue Cross Blue Shield of Massachusetts (BCBSM) Alternative Quality Contract (AQC). Results between enrollees in low and high economic status were compared.

Quality improved for all enrollees between 2007 and 2012, with a 1.2 percentage point greater improvement among the low-economic population. Improvement was greater on preventive care and pediatric process measures, but less on chronic disease-management measures.

Health outcomes improved for both populations, but changes in the disparity gap were not statistically significant. Both populations showed slower growth in spending, but again with no significant differences between the two groups (pharmaceutical spending was not included in the study).

BCBSM launched the multi-year program in 2009 with seven provider organizations. By 2012, it covered 90 percent of physicians in the BCBSM network. It uses a population health-based global budget model and rewards physicians for performance while requiring them to share financial risk.

The authors of the Commonwealth Fund-supported report noted little previous research on the impact of population-based payments on quality disparities. Some observers also noted that P4P incentives may have the potential to exacerbate disparities.

Rachel Reid, MD, associate physician policy researcher for the RAND Corporation, said the gains in process measures and the neutral results on outcomes and spending can be seen as positives.

“Getting better or staying the same is not an exacerbation of disparities,” Reid, a primary care physician at Brigham and Women’s Hospital in Boston, said in an interview. “It depends on your perspective, but not getting worse is good.”

François de Brantes, vice president and director of the Altarum Center for Payment Innovation, had a different take.

“It’s good news that they increased the number of processes of care to certain individuals who had been receiving considerably fewer. Conversely, it doesn’t seem to have changed the endpoints: quality outcomes and cost outcomes,” de Brantes said in an interview. “In baseball, one out of three is a good average, but we’re not in baseball.”

Physicians generally like process measures, de Brantes said, because they are then measured on elements that they can control.

“It’s great that patients are coming in for lab tests,” de Brantes said. “Hooray, but why is there still a gap in outcomes?”

The report noted that the study covered a commercially insured population and the lower economic status group studied had an income higher than the U.S. median. Reid said the study was “strong” but it would be interesting to see the impact of global payment models on a “less heterogeneous” population that included Medicaid enrollees.

Two-Decade Debate

Eric Schneider, MD, senior vice president for policy and research at the Commonwealth Fund, agreed on the need for a wider study. Schneider noted that the debate on whether P4P has the potential to worsen disparities has gone on for two decades and includes a May 2010 Health Affairs report he co-wrote with one of the current study’s authors.

Schneider said a larger, multi-state, multi-payer study examining the same issues would be useful—especially one with the same magnitude of financial incentives. The average P4P quality performance bonus is 2.4 percent, the study noted, but physicians in BCBSM’s AQC program in 2009-2010 could earn up to 10 percent of their risk-adjusted budgets in bonus payments.

“In general, the magnitude of incentives in pay for performance has been small,” Schneider said in an interview, which is one reason it’s often difficult to get providers to participate.

Melony Sorbero, PhD, co-director of the RAND-University of Pittsburgh Health Institute, noted also that the impact of an individual P4P program can be diluted when a physician is operating under multiple P4P contracts with different payers and with each program offering different incentives and demanding different requirements.

While the AQC incentive was significant, Sorbero said in an interview that researchers have yet to determine the two vital P4P tipping points for altering physician behavior: the size of the incentive a program offers and the percentage of the physician’s patients covered by that program.

The lack of reduction in the spending gap was not necessarily a negative finding.

“I would say this is a positive result,” Sorbero said. “They didn’t have to see a dramatic increase in spending to improve quality.”

Sorbero and Schneider both highlighted that AQC physicians were likely early adopters of quality improvement, so it’s harder for a recent intervention to produce dramatic results.

Delivery Vehicle

In this particular study, Sorbero said it was important to note that the larger P4P incentive was layered on top of a global payment arrangement.

Suzanne Delbanco, PhD, executive director for Catalyst for Payment Reform, agreed, and noted how most P4P programs are layered on top of a fee-for-service “chassis.” Delbanco said in an interview that P4P has the potential to be “a strong vehicle” for reducing disparities by spotlighting where they exist.

“But trying to perform well on a measure is not going to do all that is needed,” Delbanco said. “It takes some reengineering of how care is delivered—which is hard to do with just a pay-for-performance bonus.”

Schneider and de Brantes agreed. De Brantes urged better communication to facilitate continuous, coordinated care with follow up and follow through.

Schneider said practices that appear to have “topped out” in their quality improvement efforts need to invest in technology and be willing to redesign their operations. There also are required societal investments.

The Commonwealth Fund is studying the impact of nonmedical services on disparities, Schneider said. These include housing for the homeless, transportation services for the poor, and access to more nutritious food in low-income neighborhoods.

Focusing on these areas can improve care and lower spending, he said.

“But healthcare systems can’t pay for those services, so there’s a little bit of a policy disconnect there,” Schneider said.


Andis Robeznieks is a freelance writer based in Chicago. Follow Andis on Twitter at @AndisRobeznieks.

 

Publication Date: Friday, January 13, 2017